MSG Technical Report

8/2/2019 MSG Technical Report

1/36


2/36

2

TABLE OF CONTENTS

SUMMARY..............................................................................................................................3

1. INTRODUCTION............................................................................................................7

2. ADVERSE REACTIONS TO FOODS ..........................................................................7

2.1 Food allergies............................................................................................................8

2.2 Food intolerances ......................................................................................................9

2.3 Adverse reactions to food additives...........................................................................9

3. ADVERSE REACTIONS ATTRIBUTED TO MSG ...................................................9

3.1 Reported reactions.....................................................................................................9

3.2 Prevalence of reactions............................................................................................11 3.3 Proposed mechanisms..............................................................................................11

4. PHYSICAL AND CHEMICAL PROPERTIES OF MSG.........................................12

5. SOURCES.......................................................................................................................13

5.1 Occurrence...............................................................................................................13

5.2 Estimated intakes .....................................................................................................13

6. KINETICS AND METABOLISM ...............................................................................15

6.1 The role of glutamate in metabolism .......................................................................15

6.2 Kinetics and metabolism of dietary glutamate.........................................................16

7. REVIEW OF THE SAFETY OF MSG .......................................................................17

7.1 Previous considerations...........................................................................................17

7.2 Review of scientific literature ..................................................................................19

REFERENCES.......................................................................................................................29


3/36

3

SUMMARY

Monosodium glutamate (MSG) is the sodium salt of the non-essential amino acid glutamicacid, one of the most abundant amino acids found in nature. Glutamate is thus found in a

wide variety of foods, and in its free form has been shown to have a flavour enhancing effect.

Because of its flavour enhancing properties, glutamate is often deliberately added to foods either as the purified monosodium salt (MSG) or as hydrolysed protein.

Since the late 1960s MSG has been claimed to be the cause of a range of adverse reactions in

people who had eaten foods containing the additive. In particular, MSG has been implicated

as the causative agent in the symptom complex known as Chinese restaurant syndrome andalso as a trigger for bronchoconstriction in some asthmatic individuals.

The purpose of this report is to examine the evidence for a relationship between MSG

exposure and (i) the Chinese restaurant syndrome and (ii) the induction of an asthmatic

reaction in susceptible individuals. This assessment has considered the conclusions of

previous significant safety evaluations as well as the results of more recent studies.

Adverse reactions attributed to MSG

In the late 1960s numerous case reports appeared in the scientific literature describing acomplex of symptoms which came to be known as the Chinese restaurant syndrome (CRS)

because they typically followed ingestion of a Chinese meal. Investigations have mainlyfocussed on MSG as the causative agent in CRS. An increasing number and variety of

symptoms have been classified as CRS, however the most frequently reported symptoms are

headache, numbness/tingling, flushing, muscle tightness, and generalised weakness. Morerecently, the termMSG symptom complex has been used instead of CRS. The reports ofMSG-triggered CRS were followed in the early 1980s by reports of a possible association

between MSG and the triggering of bronchospasm/bronchoconstriction in small numbers of

asthmatics.

The prevalence of CRS is not really known but is suggested to be between 1 and 2% of thegeneral population. While a number of mechanisms have been proposed to explain how

MSG might trigger the various reported reactions, none have been proven and very little

follow-up research has been conducted to further investigate any of the proposedmechanisms.

Physical and chemical properties of MSG

MSG (MW: 187.13) is typically produced as a white crystalline powder from fermentationprocesses using molasses from sugar cane or sugar beet, as well as starch hydrolysates. MSG

has a characteristic taste called unami (savoury deliciousness), which is considered distinctfrom the four other basic tastes (sweet, sour, salty, and bitter). The optimal palatability

concentration for MSG is between 0.2 0.8% with the largest palatable dose for humans

being about 60mg/kg body weight.

Sources of MSG

Glutamate occurs naturally in virtually all foods, including meat, fish, poultry, breast milkand vegetables, with vegetables tending to contain proportionally higher levels of free


4/36

4

glutamate. Various processed and prepared foods, such as traditional seasonings, sauces andcertain restaurant foods can also contain significant levels of free glutamate, both from

natural sources and from added MSG.

No data is available on the average consumption of MSG for Australian or New Zealand

consumers however data from the United Kingdom indicates an average intake of590mg/day, with extreme users consuming as much as 2330mg/day. In a highly seasoned

restaurant meal, however, intakes as high as 5000mg or more may be possible.

Kinetics and metabolism of MSG

Glutamate occupies a central position in human metabolism. It comprises between 10 40%

by weight of most proteins, and can be synthesised in vivo. Glutamate supplies the aminogroup for the biosynthesis of all other amino acids, is a substrate for glutamine and

glutathione synthesis, is an key neurotransmitter in the brain and is also an important energy

source for certain tissues.

Humans are exposed to dietary glutamate from two main sources either from ingesteddietary protein, or ingestion of foods containing significant amounts of free glutamate (either

naturally present, or added in the form of MSG/hydrolysed protein). Dietary glutamate is

absorbed from the gut by an active transport system into mucosal cells where it ismetabolised as a significant energy source. Very little dietary glutamate actually reaches the

portal blood supply. The net effect of this is that plasma glutamate levels are onlymoderately affected by the ingestion of MSG and other dietary glutamates. Its only when

very large doses (>5g MSG as a bolus dose) are ingested, that significant increases will occur

in plasma glutamate concentration, however, even then the concentration typically returns tonormal within 2 hours. In general, foods providing metabolisable carbohydrate significantlyattenuate peak plasma glutamate levels at doses up to 150mg/kg body weight.

Breast milk concentrations of glutamate are only modestly influenced by the ingestion ofMSG and the placenta is virtually impermeable to glutamate. Although glutamate is an

important neurotransmitter in the brain, the blood brain barrier effectively excludes passiveinflux of plasma glutamate.

Review of the safety of MSG

Two major evaluations of the safety of MSG have been undertaken in recent history. The

Joint FAO/WHO Expert Committee on Food Additives (JECFA) undertook an evaluation ofMSG in 1987, and the Federation of American Societies for Experimental Biology (FASEB)

undertook a review in 1995.

The JECFA and FASEB reviews both concluded that MSG does not represent a hazard tohealth for the general population. In relation to MSG being a cause of adverse effects in a

subset of the population the two expert bodies reached slightly differing conclusions.


5/36

5

JECFA noted that controlled double-blind crossover trials have failed to demonstrate anunequivocal relationship between CRS and consumption of MSG and also that MSG has not

been shown to provoke bronchoconstriction in asthmatics. The FASEB evaluation concludedthat sufficient evidence exists to indicate some individuals may experience manifestations of

CRS when exposed to a 3g bolus dose of MSG in the absence of food. In addition, they

concluded there may be a small number of unstable asthmatics who respond to doses of 1.5 2.5g of MSG in the absence of food.

In reviewing the individual studies considered by both the JECFA and FASEB evaluations as

well as more recent studies it is clear that many of the earlier studies have suffered from

numerous methodological flaws and have produced conflicting and inconclusive results,which are difficult to reconcile. The more recent studies those conducted following the

FASEB review have largely addressed many of the earlier study design problems and theirresults may thus be considered more reliable.

In relation to more serious adverse effects, the bulk of the clinical and scientific investigation

has focussed on the triggering of asthmatic attacks. The evidence for MSG as a cause of suchreactions however is inconclusive. The more recently conducted studies, which wereundertaken with asthmatic individuals who believed themselves to be sensitive to MSG,

would suggest that MSG is not a significant trigger factor. Follow up studies would be

helpful to confirm this finding.

In relation to CRS, the evidence from recent studies supports the conclusions reached in theFASEB review. Namely, that ingestion of large amounts (3g) of MSG in the absence of

food may be responsible for provoking symptoms similar to CRS in a small subset of

individuals. These symptoms, although unpleasant, are neither persistent nor serious. AsMSG would always be consumed in the presence of food, an important question that remainsunanswered by the scientific literature is what effect consumption with food would have on

the incidence and severity of symptoms. The pharmacokinetic evidence suggests food,

particularly carbohydrate, would have an attenuating affect.

Although the prevalence of CRS has been estimated to be about 1 2% of the generalpopulation it is not clear what proportion of the reactions, if any, can be attributed to MSG.

The vast majority of reports of CRS are anecdotal, and are not linked to the actual glutamate

content of the food consumed. Furthermore, when individuals with a suspected sensitivity toMSG are tested in double-blind challenges the majority do not react to MSG under the

conditions of the study (or react equally to placebo). Many individuals may therefore

incorrectly be ascribing various symptoms to MSG, when in fact some other food componentmay be the cause. This highlights the need for individuals with suspected MSG sensitivity to

undergo appropriate clinical testing.

While many of the more recently conducted studies have addressed the design flaws of earlierstudies, one of the difficulties remaining is that the CRS symptoms are highly subjective in

nature and are rarely associated with any objective clinical signs (e.g. vomiting, increased

pulse rate, etc). The placebo response therefore plays a significant role in many of thereactions observed, making it difficult to interpret the significance of any responses to MSG.

The elucidation of a possible mechanism of CRS, plus associated objective clinical measures,would greatly aid in the further study of this symptom complex.


6/36

6

Conclusion

There is no convincing evidence that MSG is a significant factor in causing systemicreactions resulting in severe illness or mortality. The studies conducted to date on CRS have

largely failed to demonstrate a causal association with MSG. Symptoms resembling those of

CRS may be provoked in a clinical setting in small numbers of individuals by theadministration of large doses of MSG without food. However, such affects are neither

persistent nor serious and are likely to be attenuated when MSG is consumed with food. Interms of more serious adverse effects such as the triggering of bronchospasm in asthmatic

individuals, the evidence does not indicate that MSG is a significant trigger factor.


7/36

7

1. INTRODUCTION

Monosodium glutamate (MSG) is the sodium salt of the non-essential amino acid glutamicacid. Glutamic acid is one of the most abundant amino acids found in nature and exists both

as free glutamate and bound with other amino acids into protein. Animal proteins may

contain about 11 to 22% by weight of glutamic acid, with plant proteins containing as muchas 40% glutamate (Giacometti 1979). Glutamate is thus found in a wide variety of foods, and

in its free form, where it has been shown to have a flavour enhancing effect, is also present inrelatively high concentrations is some foods such as tomatoes, mushrooms, peas and certain

cheeses. As a result of its flavour enhancing effects, glutamate is often deliberately added to

foods either as the purified monosodium salt (MSG) or as a component of a mix of aminoacids and small peptides resulting from the acid or enzymatic hydrolysis of proteins (e.g.

hydrolysed vegetable protein or HVP). Other substances, such as sodium caseinate andnatural flavourings, are also added to many savoury foods and these can also contain

considerable amounts of free glutamate.

The use of added MSG became controversial in the late 1960s when it was claimed to be thecause of a range of adverse reactions in people who had eaten foods containing the additive.An ongoing debate exists as to whether MSG in fact causes any of these symptoms and, if so,

the prevalence of reactions to MSG.

The purpose of this assessment is to review previous considerations of the safety of MSG, as

well as any more recent scientific publications, to determine if MSG has the potential tocause severe adverse reactions when ingested with food.

2. ADVERSE REACTIONS TO FOODS

Adverse reactions to food can be defined as any abnormal physiological response to a

particular food (Taylor 2000) and can be classified into a number of different categories of

reaction (Wthrich 1996), as illustrated below.

Adverse Food Reactions

Toxic Reactions Hypersensi tivity Reactions

Food Allergies Food Intolerances

ImmediateHypersensitive

Reactions

DelayedHypersensitive

Reactions

Toxic reactions will occur in virtually all individuals in a dose-dependent manner, whereashypersensitivity reactions are usually idiosyncratic reactions that only occur in a small subset


8/36

8

of individuals. Hypersensitivity reactions can be further divided into two majorsubcategories food allergies and food intolerances. Food allergies are immune system-

mediated and can be classified as either immediate or delayed hypersensitivity reactionswhereas food intolerances are non-immune system-mediated.

2.1 Food allergies

Food allergies are an abnormal response by the bodys immune system to certain componentsof foods, usually specific proteins. True food allergies may involve several types of

immunological responses (Sampson and Burks 1996). The most common food allergy

reactions are the immediate hypersensitivity reactions, which are mediated by allergen-specific immunoglobulin E (IgE) antibodies. Symptoms of IgE-mediated allergic reactions,

such as acute urticaria or anaphylaxis, can occur immediately after ingestion of the offendingfood, depending on the dose ingested but they may be delayed by several hours in other

cases, such as atopic dermatitis.

Although all humans have low levels of circulating IgE antibodies, only individualspredisposed to the development of allergies produce IgE antibodies that are specific for andrecognise allergens. The IgE-mediated response is divided into two stages: (i) sensitisation;

and (ii) the allergic reaction. Exposure to a food allergen elicits the formation of specific IgE

antibodies by the B-lymphocytes. The IgE antibodies attach with exceptionally high affinityto receptors on the surface of tissue mast cells and blood basophils (immature red blood

cells). At this point the individual is sensitised to the allergenic substance but has yet toexperience an allergic reaction. Subsequent exposure to the allergen will result in the cross-

linking of the allergen to the IgE molecules on the mast/basophil cell surface. The cross-

linking triggers the mast/basophil cells to release various chemical mediators, such ashistamine and cytokines. The release of these mediators results in various inflammatoryreactions that may occur in the skin, gastrointestinal tract or the respiratory tract. In extreme

cases, food allergens can cause anaphylactic shock resulting in the rapid and potentially life

threatening collapse of the cardio-respiratory system.

IgE-mediated food allergies affect between 1 and 2% of the population (Metcalfe et al1996,Niestijl-Jansen et al1994), however, infants and young children are more commonly affected

with the prevalence in children under three years of age being between 5 and 8% (Bock 1987,

Sampson 1990a, Tayloret al1989).

True food allergies also include delayed hypersensitivity reactions, the mechanisms of which

are less clear. Such reactions include cell-mediated mechanisms involving sensitisedlymphocytes in tissues, rather than antibodies (Sampson 1990b). In cell-mediated reactions,

the onset of symptoms occurs more than 8 hours after ingestion of the offending food. Theprevalence of food-induced, cell-mediated reactions is not known (Burks and Sampson 1993)

but the reactions are well documented in infants and typically occur following exposure tomilk and soybeans. The most common cell-mediated hypersensitivity reaction affecting all

age groups is coeliac disease, also known as gluten-sensitive enteropathy. Coeliac disease

results from an abnormal response of the T lymphocytes in the small intestine to the glutenproteins in cereals and affects genetically predisposed individuals. The T cells have specific

markers on their surface that recognise the allergen deposited at a local site such as thegastrointestinal mucous membrane, resulting in an inflammatory reaction affecting the

epithelium of the small intestine.


9/36

9

2.2 Food intolerances

Food intolerances can be described as any form of food sensitivity that does not involve animmunological mechanism. They can be classified according to their mechanism e.g.,

enzymatic, pharmacological or undefined (Wthrich 1996, Anderson 1996), or alternatively

can be defined in terms of the reactions they elicit e.g., metabolic food disorders,anaphylactoid reactions or idiosyncratic reactions (Taylor 2000). Food intolerances usually

produce less severe symptoms than food allergies, and affected individuals can usuallytolerate some of the offending food in their diets.

The best-known examples of metabolic food disorders are lactose intolerance and favismboth of which involve the inherited deficiency of an enzyme. In the case of lactose

intolerance the reaction is due to an inherited deficiency of the enzyme lactase in the gut ofthe affected persons. Favism is intolerance to consumption of faba beans or inhalation of

pollen from the Vicia faba plant. Reactions are due to an inherited deficiency of the enzyme,

erythrocyte glucose-6-phosphate dehydrogenase. Most metabolic food disorders are

genetically acquired and both lactose intolerance and favism occur at much higherfrequencies in certain ethnic groups (Taylor 2000).

Anaphylactoid reactions have symptoms similar to those of anaphylaxis, but are triggered

instead by non-immunological mechanisms, which directly lead to the release of chemicalmediators from mast cells. To date, no specific substances in foods causing this response

have been identified, with the majority of cases being associated with the administration ofcertain drugs or the radio-contrast dyes used for X-ray studies.

Idiosyncratic reactions refer to adverse reactions where the mechanism is undefined. Oneexample is sulphite-induced asthma, which has been estimated to affect 1 2% of allasthmatics.

2.3 Adverse reactions to food additives

Sensitivity to most food additives is believed to occur in only a small minority of thepopulation (ANZFA 1997, MAFF 1987), with most adverse effects due to various

pharmacological and other non-immunological mechanisms (Hannuksela and Haahtela

1987), rather than being true allergic reactions.

Exacerbation of asthma is one of the adverse effects most typically reported as being

associated with food additives. Although 23 to 67% of people with asthma perceive that foodadditives exacerbate their asthma (Dawson et al1990, Abramson et al1995), various double

blind, placebo-controlled trials report a prevalence rate of less than 5% (Bock and Aitkins1990, Onorato et al1986).

3. ADVERSE REACTIONS ATTRIBUTED TO MSG

3.1 Reported reactions

In 1968, a letter was published in theNew England Journal of Medicine describing asyndrome, which began 15 to 30 minutes after eating in certain Chinese restaurants, and

lasted about 2 hours with no lasting effects. The symptoms were described as numbness atthe back of the neck, gradually radiating to both arms and the back, general weakness and


10/36

10

palpitation (Kwok 1968). The author noted that the symptoms simulated those he has hadfrom hypersensitivity to acetylsalicylic acid, but were milder. The author suggested

numerous possible causes for the symptoms, including alcohol, salt and MSG used incooking. The term Chinese Restaurant Syndrome (CRS) was coined to describe the

symptom complex.

Since that time numerous other case reports have appeared in the literature, with the focus

mainly on MSG as the causative agent in CRS. An increasing number and variety ofsymptoms have also subsequently been added to the list of manifestations of CRS. In 1995,

the Federation of American Societies for Experimental Biology (FASEB), who had been

commissioned by the United States Food and Drug Administration (FDA) to undertake areview of reported adverse reactions to MSG, reported that the following symptoms are

considered representative of the acute, temporary, and self-limited reactions to oral ingestionof MSG (FASEB 1995):

- burning sensations in the back of the neck, forearms, chest;

- facial pressure/tightness;- chest pain;- headache;

- nausea;

- palpitation;- numbness in back of neck, radiating to arms and back;

- tingling, warmth, weakness in face, temples, upper back, neck and arms;- bronchospasm (observed in asthmatics only);

- drowsiness;

- weakness.

In its report, FASEB noted that this catalogue of symptoms is based on testimonial reports

received by the FDA Adverse Reaction Monitoring System as well as a review of the

literature and is therefore based on accounts that are anecdotal and not verifiable. TheFASEB report indicated that while the testimonial reports do not establish causality by MSG,

the overall impression of the Expert Panel was that causality had been demonstrated.

Reports of more serious symptoms, such as atrial fibrillation, ventricular tachycardia and

arrhythmias were not given any credence by the FASEB, as they were single case reports thatlacked confirmatory evidence linking the reactions to MSG content of foods (Raiten et al

1995).

In the FASEB report, the term Chinese restaurant syndrome was abandoned as pejorative,

and instead the termMSG symptom complex was used to describe the range of symptomsexperienced by affected individuals.

An interesting feature of the CRS is that the presentation of symptoms often varies, with

affected individuals usually only reporting one or a few of the characterising symptoms at

any one time. In some recently conducted studies, the most frequently reported symptomswere headache, numbness/tingling, flushing, muscle tightness, and generalised weakness

(Yang et al1997, Geha et al2000a).


11/36

11

3.2 Prevalence of reactions

A small number of studies have been conducted to try and determine the true prevalence ofCRS and these have produced conflicting results. While one survey has classified CRS as

very common, putting its prevalence at 25% (Reif-Lehrer 1977), another survey has

estimated its prevalence to be much lower, at between 1 to 2% of the general population(Kerret al1979a). The conflicting results appear in part to be due to the way the studies

have been conducted and also the way various symptoms have been characterised by thedifferent investigators.

The Reif-Lehrer (1977) survey, which estimated the prevalence of reactions to be 25%, hasbeen criticised as having several inherent biases and therefore is considered to represent an

exaggerated estimate of the true prevalence (Kerret al1979b, Pulce 1992, Geha et al2000b).The main criticisms relate to methodological problems, such as demand bias in the

questionnaire where leading questions such as Do you think you get Chinese restaurant

syndrome? were asked, and population bias, where the surveyed population was not

considered representative of the general population and had a higher than average awarenessof CRS prior to the survey. Another major criticism is that the clinical criteria used forselecting reactors from non-reactors were quite broad and thus could have lead to an

overestimate of CRS prevalence in the population group studied.

A slightly later survey by Kerret al(1979a), which reported an estimated prevalence for

possible CRS of between 1 and 2%, attempted to redress some of the biases inherent in thefirst survey, and thus is considered a more reliable indicator of the true prevalence of

reactions. This survey was conducted using the National Consumer Panel of the Market

Research Corporation of America, and therefore should have avoided any population bias.Efforts were also made to avoid demand-biased questions in the questionnaires used. Kerretal(1979b) noted however that many unresolved issues still remain in relation to the true

prevalence of CRS. The most problematic of these is that numerous symptoms have been

associated with CRS and many of these symptoms are ambiguous and imprecise. Thevarious clinical presentations thus make it difficult to accurately diagnose CRS and this is

likely an important confounding factor in questionnaire surveys.

3.3 Proposed mechanisms

Numerous mechanisms have been proposed for CRS. While some of the proposed

mechanisms postulate an involvement for MSG, others do not.

It has been suggested that CRS resembles an immediate hypersensitivity reaction in that the

symptoms typically occur within a few minutes to several hours after eating the offendingfood. However, no evidence for an IgE-mediated reaction exists (Pulce et al1992), although

the possibility of an anaphylactoid reaction cannot be discounted. Other non-allergenicmechanisms that have been suggested as the cause of CRS include acetylcholinosis, vitamin

B6 deficiency, reflux oesophagitis, and histamine toxicity.

Ghadimi et al(1971) suggested that CRS was the result of an increase in acetylcholine

caused by the ingestion of MSG in large doses with the glutamate being converted toacetylcholine via the tricarboxylic acid (TCA) cycle. A similarity between the symptoms of

CRS and those occurring after injection of acetylcholine (flushing, feeling of warmth,throbbing in the head, palpitations, and substernal constriction) was noted and it has also been


12/36

12

observed experimentally that in humans there is a 28% decrease in cholinesterase after MSGis ingested. The symptoms of CRS were also found to be capable of modulation using drugs

affecting the cholinergic mechanisms.

Folkers et al(1984) have suggested that the reactions experienced by MSG-sensitive

individuals are a result of vitamin B6 deficiency. They found that when MSG respondersreceived supplemental B6, CRS symptoms were prevented.

Kenney (1986) has suggested that the symptoms seen in CRS are caused by MSG but are not

a neurological/physiological reaction. He has suggested that CRS is actually a case of reflux

oesophagitis, with MSG acting as an oesophageal irritant. The symptoms and regions of thebody affected by CRS were noted to be similar to those of pain referred from the upper

oesophagus. Studies have shown that a variety of seemingly unrelated substances such ascoffee, orange juice and tomato juice, ingested via oesophageal infusion, can cause similar

types of symptoms (Price et al1978). Adding weight to this hypothesis are the results of

studies suggesting that individuals reacting to MSG may react to concentration rather than

dose and that the same dose taken in capsules is associated with fewer reactions.

Chin et al(1989) suggested that there are similarities between CRS and scombroid poisoning,

caused by naturally occurring histamine in foods and they therefore undertook assays of

several common Chinese restaurant dishes and condiments for histamine content. It wasconcluded that while the histamine content of most of the foods assayed was not sufficient

alone to cause histamine toxicity, in certain situations histamine intake over the course of anentire meal could approach toxic levels.

To date, very little research has been done to investigate any of these proposed mechanismsfurther. The FASEB report (1995) found that a major constraint in identifying mechanismshas been the inability to make connections between studies of adverse effects and those of

metabolic response to oral MSG challenges. The former lack data on any objective measures

of response, in particular, blood glutamate concentrations, and the latter focus on bloodglutamate data without evaluation of adverse effects.

4. PHYSICAL AND CHEMICAL PROPERTIES OF MSG

MSG (MW: 187.13) is typically marketed as a white crystalline powder and is readily solublein water but sparingly soluble in ethanol. MSG is not hygroscopic and is considered quite

stable in that it does not change in appearance or quality during prolonged storage at room

temperature. MSG does not decompose during normal food processing or cooking but inacidic conditions (pH 2.2-2.4) and at high temperatures it is partially dehydrated and

converted into 5-pyrrolidone-2-carboxylate (Yamaguchi and Ninomiya 1998). The chemicalstructure of MSG is shown in Figure 2 below.


13/36

13

Figure 2: Chemical structure of MSG

NaO

NH2

OH

OO

MSG is produced today through fermentation processes using molasses from sugar cane or

sugar beet, as well as starch hydrolysates from corn, tapioca etc. Prior to the development ofthe fermentation process, MSG was produced by hydrolysis of natural proteins, such as wheat

gluten and defatted soybean flakes.

MSG is a taste active chemical and is said to impart a unique taste. The characteristic taste of

MSG is a function of its stereochemical structure with the D-isomer having no characteristictaste. The MSG taste is readily identified in Asian cultures as being distinct from the four

basic tastes (sweet, sour, salty, bitter) and has been called unami. Roughly translated, unamimeans savoury deliciousness. Western cultures have had difficulty in describing this taste

and thus have not identified it as unique. More recently however unami has gained

widespread acceptance as a fifth basic taste (Yamaguchi and Ninomiya 2000).

The optimal palatability concentration for MSG is between 0.2 0.8% and its use tends to beself-limiting as over-use decreases palatability. The largest palatable dose for humans is

about 60mg/kg body weight (Walker and Lupien 2000).

5. SOURCES

5.1 Occurrence

As an abundant amino acid, glutamate is found in a virtually all foods, including meat, fish,poultry, breast milk and vegetables. In general, protein-rich foods such as breast milk, cheese

and meat, contain large amounts of bound glutamate, while most vegetables contain relatively

low amounts. However, despite their lower protein contents, vegetables tend to containproportionally higher levels of free glutamate, especially peas, tomatoes, and potatoes. The

typical glutamate content of various foods is given in Table 1. The free glutamate content ofother foods such as traditional seasonings, packaged foods and restaurant food is presented in

Table 2.

5.2 Estimated intakes

There is no data available on the average consumption of MSG for Australian or NewZealand consumers. Data from the United Kingdom indicates an average intake of

590mg/day, with extreme users (97.5th

percentile consumers) consuming 2330mg/day

(Rhodes et al1991). In a highly seasoned restaurant meal, however, intakes as high as

5000mg or more may be possible (Yang et al1997).


14/36

14

Table 1: Naturally occurring glutamate in various foods

Food Bound glutamate

(mg/100g)

Free glutamate

(mg/100g)

Milk/dairy products:

Cows milk

Human milk

Parmesan cheese

819

229

9847

2

22

1200

Poultry products:

Eggs

Chicken

Duck

1583

3309

3636

23

44

69

Meat:

Beef

Pork

2846

2325

33

23

Fish:

Cod

Mackerel

Salmon

2101

2382

2216

9

36

20Vegetables:

Peas

Corn

Carrots

Spinach

Tomatoes

Potato

5583

1765

218

289

238

280

200

130

33

39

140

180Source: Yamaguchi and Ninomiya 1998

Table 2: Free glutamate content of traditional seasonings, various

packaged foods and restaurant meals

Food type Free glutamate content(mg/100g)

Concentrated extracts:

Vegemite

Marmite

Oyster sauce

1431

1960

900

Soy sauce:

China

Japan

Korea

Phillippines

926

782

1264

412

Fish sauce:Nam-pla

Nuoc-mam

Ishiru

Bakasang

950

950

1383

727

Condensed soups 0 480

Sauces, mixes, seasonings 20 1900

Chinese restaurant meals


15/36

15

6. KINETICS AND METABOLISM

6.1 The role of glutamate in metabolism

Glutamate performs a myriad of essential roles in intermediary metabolism and is present in

large amounts in the organs and tissues of the body. The daily turnover of glutamate in theadult human has been estimated as 4800mg (Munro 1979). Some of the important metabolic

roles of glutamate include:

A substrate for protein synthesis as one of the most abundant amino acids present in

nature, comprising between 10 40% by weight of most proteins, L-glutamic acid isan essential substrate for protein synthesis. Glutamic acid possesses physical and

chemical characteristics which make it a principal contributor to the secondarystructure of proteins, namely the -helices (Young and Ajami 2000);

A transamination partner with -ketoglutarate L-glutamate is synthesised from

ammonia and -ketoglutarate (an intermediate of the citric acid cycle) in a reactioncatalysed by L-glutamate dehydrogenase. This reaction is of fundamental importancein the biosynthesis of all amino acids, since glutamate is the amino group donor in the

biosynthesis of other amino acids through transamination reactions (Lehninger 1982);

A precursor of glutamine glutamine is formed from glutamate by the action of

glutamine synthetase. This is also an important central reaction in amino acidmetabolism since it is the main pathway for converting free ammonia into glutamine

for transport in the blood. Glutamate and glutamine are thus key links between

carbon and nitrogen metabolism in general and between the carbon metabolism ofcarbohydrate and protein in particular (Reeds et al2000);

A substrate for glutathione production glutathione, a tripeptide composed of

glutamic acid, cysteine and glycine, is present in all animal cells and serves as areductant of toxic peroxides by the action of glutathione peroxidase. Glutathione is

also postulated to function in the transport of amino acids across cell membranes(Lehninger 1982);

A precursor ofN-acetylglutamate an essential allosteric activator of carbamylphosphate synthetase I, a key regulatory enzyme in the urea cycle, ensuring that the

rate of urea synthesis is in accord with rates of amino acid deamination (Brosnan

2000);

An important neurotransmitter glutamate is the major excitatory transmitter withinthe brain, mediating fast synaptic transmission and is active in perhaps one third of

central nervous system synapses (Watkins and Evans 1981). Glutamate is also aprecursor to another neurotransmitter GABA;

An important energy source for some tissues (mucosa) intestinal tissues areresponsible for significant metabolism of dietary glutamate, where it serves as a

significant energy yielding substrate (Young and Ajami 2000). A net effect of theextensive intestinal metabolism of dietary glutamate is a relatively stable plasma

glutamate concentration throughout fasting and fed periods.


16/36

16

6.2 Kinetics and metabolism of dietary glutamate

Humans are exposed to dietary glutamate from two main sources either from the digestionof ingested dietary protein, or from the ingestion of foods that contain significant amounts of

free glutamate (either naturally present, or added in the form of MSG/hydrolysed protein).

Glutamate is absorbed from the gut by an active transport system specific for amino acids.

This process is saturable, can be competitively inhibited and is dependent on sodium ionconcentration (Schultz et al1970). Glutamic acid in dietary protein is digested to free amino

acids and small peptides, both of which are absorbed into mucosal cells where peptides are

hydrolysed to free amino acids and some of the glutamate is metabolised. Excess glutamateappears in the portal blood, where it is metabolised by the liver.

A number of early studies with dogs (Neame and Wiseman 1958), and later, studies

conducted in rats (Windmueller 1982, Windmueller & Spaeth 1974, 1975), demonstrated that

the vast majority of dietary glutamate is metabolised by the gastrointestinal tract. In fact,

very little dietary glutamate enters either the systemic or the portal blood supply (Young andAjami et al2000), indicating it is almost exclusively utilised by the intestinal tissues.

The process of dietary glutamate utilisation by the intestinal tract has recently been

extensively studied using enteral infusions of [13

C5] glutamate in rapidly growing pigletsconsuming diets based on whole-milk proteins (Reeds et al1996, 1997, 2000). The results

showed that 95% of dietary glutamate presented to the mucosa was metabolised in first passand that of this, 50% appeared as portal CO2, with lesser amounts as lactate and alanine. This

indicates that glutamate is the single largest contributor to intestinal energy generation. The

studies also indicated that about 10% of dietary glutamate is incorporated into mucosalprotein synthesis, with the remainder being used for the synthesis of proline, arginine andglutathione. In fact, all three substances proline, arginine and glutathione are derived

almost exclusively from dietary glutamate, rather than the vast in vivo pool of glutamate.

As a consequence of the rapid metabolism of glutamate in intestinal mucosal cells, with any

excess glutamate being metabolised by the liver, systemic plasma levels are typically low,even after ingestion of large amounts of dietary protein (Munro 1979, Meister 1979). Human

plasma is reported to contain between 4.4 8.8 mg/L of free glutamate (Pulce et al1992).

Studies on the effects of food on glutamate absorption and plasma levels have been done in

mice, pigs and monkeys as well as humans. When infant mice were given MSG with infant

formula or when adults were given MSG with consomm by gastric intubation, peak plasmaglutamate levels were markedly lower than when the same dose was given in water, with the

time to reach peak levels being longer (Ohara et al1977). Similar effects of food onglutamate absorption and plasma levels have been observed in humans. Only slight rises in

plasma glutamate have been observed following ingestion of a dose of 150 mg/kg bw toadults with a meal, with human infants, including premature babies, also demonstrating the

same capacity to metabolise similar doses given in infant formula (Tung and Tung 1980).

Human plasma glutamate levels were much lower when large doses of MSG were ingestedwith meals compared to ingestion in water. In general, foods providing metabolisable

carbohydrate significantly attenuate peak plasma glutamate levels at doses up to 150mg/kgbody weight (Bizzi et al1977, Steginket al1979a, 1979b, 1982, 1983a, 1983b, 1983c, 1985,

1986).


17/36

17

In reviewing all the evidence in relation to the effect of MSG ingestion on plasma glutamatelevels, the FASEB Expert Panel concluded that the composition of the dosing vehicle as well

as the conditions of administration of the dose can significantly impact on changes incirculating glutamate in response to oral ingestion (Raiten et al1995). Overall, the evidence

indicates that the extent of the rise in plasma concentrations of glutamate is affected by a

number of factors including the size of the dose (increases with increasing dose); the natureof the dosing vehicle (e.g. water causes greater rise than a mixed meal); the temporal

proximity of food consumption (fasted subjects exhibit a greater response than those dosedwith a meal); and macronutrient composition of the concurrent food (carbohydrate and mixed

meals have an attenuating effect compared with fasting or protein).

Breast milk concentrations of glutamate are quite high and are also influenced only modestly

by the ingestion of MSG (Pitkin et al1979, Steginket al1972). Of the twenty free aminoacids in human breast milk, glutamate is the most abundant, accounting for >50% of the total

free amino acid content (Rassin et al1978). Up to 540mg glutamate/L has been found in

human milk, whereas cows milk contains 10-20mg/L (Ninomiya 1998).

The placenta is considered virtually impermeable to glutamate (Battaglia 2000). Studieswith both sheep and humans have shown the placenta removes glutamate from foetal

circulation, while concurrently supplying glutamine into the foetal circulation in very large

amounts (Lemons et al1976, Hayashi et al1978).

Although glutamate is an important neurotransmitter in the brain, the blood brain barriereffectively excludes passive influx of plasma glutamate. In guinea pigs, rats and mice, brain

glutamate levels remained unchanged after administration of large oral doses of MSG which

resulted in plasma levels increasing up to 18-fold (Peng et al1973, Liebschultz et al1977,Caccia et al1982, Airoldi et al1979, Bizzi et al1977). Brain glutamate increasedsignificantly only when plasma levels were about 20 times basal values following an oral

dose of 2g MSG/kg body weight (Bizzi et al1977). The majority of the glutamate used by

the brain is derived from local synthesis from glutamine and TCA cycle intermediates and aconsiderable fraction is also derived from the recycling of brain protein (Smith 2000).

7. REVIEW OF THE SAFETY OF MSG

7.1 Previous considerations

7.1.1 JECFA safety evaluations

The Joint FAO/WHO Expert Committee on Food Additives (JECFA) has undertaken two

evaluations of the safety of MSG. The first of these was conducted in 1971 1974, and thesecond was conducted in 1987. This review will consider only the most recent evaluation

(JECFA 1988).

JECFA examined acute, subchronic, and chronic toxicity studies in rats, mice and dogs,

together with studies on reproductive toxicity and teratology. Glutamate was found to have avery low acute oral toxicity. The LD50 for rats and mice is about 15,000 and 18,000mg/kg

body weight, respectively. Subchronic studies as well as chronic studies of up to two yearsduration in mice and rats, including a reproductive phase, did not reveal any specific adverse

effects at dietary levels of up to 4%. A two-year study in dogs at dietary levels of 10% alsodid not reveal any effects on weight gain, organ weights, clinical indices, mortality or general


18/36

18

behaviour. Reproduction and teratology studies using the oral route of administration did notreveal any adverse effects, even at high doses.

The JECFA evaluation also addressed two other issues. These were (i) potential

neurotoxicity, especially to the infant, and (ii) the putative role of MSG in CRS.

(i) Potential neurotoxicity

Examination of potential neurotoxicity was a major component of the safety evaluation, with

reports from 59 separate studies in mice, rats, hamsters, dogs, rabbits, guinea pigs, duck and

primates being considered. This issue was given a large amount of attention because ofreports that lesions (focal necrosis) in the hypothalamus were observed reproducibly in

rodents and rabbits after intravenous or subcutaneous administration of glutamate or aftervery high bolus doses by gavage. The neural lesions were observed within hours of

administration and the mouse appeared to be the most sensitive species. Notably, most of the

studies with primates were negative with regard to hypothalamic lesions.

The oral gavage doses required to produce the lesions were of the order of 1000mg/kg bodyweight as a bolus dose. The threshold blood levels associated with neuronal damage in the

mouse are 100 300mol/dL in neonates rising to 380mol/dL in weanlings and >

630mol/dL in adult mice. In humans, plasma levels of this magnitude have not beenrecorded even after bolus doses of 150mg/kg body weight (about 10g for an adult). The oral

ED50 for production of hypothalamic lesions in the neonatal mouse is about 500mg/kg bodyweight by gavage, whereas the largest palatable dose for humans is about 60mg/kg body

weight with higher doses causing nausea. It was thus concluded that voluntary ingestion

would not exceed this level.

(ii) Putative role of MSG in CRS

In consideration of idiosyncratic intolerance to MSG, most of the reports of reactions werefound to be anecdotal, however a number of studies that had been undertaken with human

volunteers were reviewed. Examination of these studies failed to demonstrate that MSG wasthe causal agent in provoking the full range of symptoms associated with CRS. It was

therefore concluded that controlled double-blind crossover trials have failed to demonstrate

an unequivocal relationship between CRS and consumption of MSG and also that MSG hasnot been shown to provoke bronchoconstriction in asthmatics.

It was concluded that the total dietary intake of glutamates arising from their use at levelsnecessary to achieve the desired technological effects and from their acceptable background

in food do not represent a hazard to health. For that reason, the establishment of anAcceptable Daily Intake (ADI) was not considered necessary, and an ADI not specified

was allocated to L-glutamic acid and the monosodium, potassium, calcium and ammoniumsalts.

It was also noted that the available evidence did not indicate that pregnant women and infantswere at any greater risk in relation to exposure to glutamate than other members of the

general population.


19/36

19

7.1.2 FASEB review

In response to continuing reports of adverse reactions to MSG and other glutamate-containingingredients, the United States FDA contracted the FASEB to conduct a review of reported

adverse reactions to MSG. The full report of the study was released in 1995 (FASEB 1995).

The report concluded that, although there was no scientifically verifiable evidence of adverse

effects in most individuals exposed to high levels of MSG, there is sufficient documentaryevidence to indicate there is a subgroup of presumably healthy individuals that responds,

generally within 1 hour of exposure, with manifestations of the MSG symptom complex

when exposed to an oral (bolus) dose of MSG of 3g in the absence of food. The report alsostated available data suggest strongly that ingestion of MSG in capsule form on an empty

stomach is more often associated with occurrence of adverse reactions, than is ingestion withfood.

In relation to asthma, the report concluded that the only scientifically verified adverse effects

of MSG in humans that have been reported are initiations of bronchospasms in a subgroup ofpeople with severe unstable asthma. The report stated that there appears to be a small subsetof people with severe unstable asthma who respond to doses of 1.5-2.5g of MSG given in a

low energy challenge vehicle e.g. a capsule, in the absence of a meal containing protein and

carbohydrate.

The report recommended that to confirm the MSG symptom complex, multiple double blind,placebo-controlled challenges on separate occasions must reproduce symptoms with the

ingestion of MSG and produce no response with placebo. The Expert Panel suggested that

five separated challenges would be necessary to conclude that subjective symptoms (e.g.headache, chest tightness, numbness, etc) are secondary to MSG in highly suggestibleindividuals, whereas only three would be necessary for those individuals not considered

highly suggestible. In individuals with objective findings (e.g. bronchospasm, vomiting etc),

a single double blind challenge was considered sufficient. The Expert Panel recognised thatthe use of capsules ensures the greatest control over dose and blinding, however, they also

noted that the use of capsules obviates the potential role of the oral cavity and oesophagus inthe precipitation of potential adverse effects. The Expert Panel suggested that the use of

capsules versus liquids would depend on the goal of the study. For example, if the goal is to

study the potential for adverse effects of MSG ingestion under conditions of normal use, aliquid vehicle would be most appropriate. The Expert Panel also noted the results of a study

by Steginket al(1979b) where administration of MSG in capsules resulted in a 3 to 4-fold

attenuation of peak plasma glutamate levels.

7.2 Review of scientific literature

7.2.1 MSG as a trigger factor for asthmatic attacks

Asthma is a relatively common disorder that can have serious consequences for the sufferer,

including death and therefore is a significant public health problem. In Australia, asthmaaffects between 22 24% of children and 13% of adults (Robertson et al1991, Abramson et

al1992), although the prevalence of food-induced asthma is somewhat lower and has beenestimated to affect 0.24% of adults and 11% of children (Woods 1997).


20/36

20

The causes of asthma are complicated and can vary from patient to patient, howeverinflammation of the bronchial airways is the characteristic finding in the majority of

asthmatic patients (OByrne 1997). Multiple trigger factors can activate asthma attacks inasthmatic patients already afflicted with inflammation of the bronchial tree and these factors

will vary from patient to patient but are important because identification and avoidance of

such trigger factors can substantially improve the quality of life of asthmatic individuals(Stevenson 2000).

A possible association between MSG and the triggering of asthma attacks was first suggested

in 1981 (Allen and Baker 1981). Since then a small number of studies have been conducted

to investigate this association but have produced conflicting results. Five of these studies didnot demonstrate MSG-induced asthma attacks (Schwartzstein et al1987, Germano et al1991,

Altman et al1994, Woods et al1998, Woessneret al1999), whereas three have concludedthat some people with asthma do get MSG-induced attacks (Allen et al1987, Moneret-

Vautrin 1987, Hodge et al1996).

The study by Allen et al(1987) recruited 32 subjects, including two subjects who were thesubject of the original case report (Allen and Baker 1981). Of the 32 who were studied, 14gave a history of asthmatic attacks after consuming a Chinese meal, with the other 18 having

unstable asthma and a reported sensitivity to other chemicals (aspirin, benzoic acid,

tartrazine, and sulphites). All subjects underwent single blind oral challenges with MSG (0.5,1.5, and 2.5g in capsules) followed by peak expiratory flow (PEF) measurements for 12 hours

after each challenge. PEF measures how fast a subject can blow air out of their lungs. Apositive response was defined as a 20% decline in PEF. Some of the challenges were

conducted in the morning and some in the afternoon. Subjects followed a specific exclusion

diet (specific details not provided) beginning 5 days before challenges. Some asthmamedications (theophylline) were ceased prior to the challenges. One subject was reported toreact to all three doses, another to the 1.5g dose only and 12 to 2.5g only. Thirteen subjects

were thus concluded to have experienced an MSG-induced asthma attack.

This study has been criticised for a variety of reasons, including: a lack of blinding of

observers, that is, the study used a single blind, rather than a double blind protocol;inadequate procedures for establishing baseline and control data; the use of effort-dependent

PEF, which can be influenced by subject bias; the cessation of anti-inflammatory and

bronchodilator medications just prior to the challenge sequence making it hard to judgewhether an asthmatic attack is due to the challenge substance, rather than simply a result of

the withdrawal of therapy; and no measurements of immunologic inflammatory markers or

changes in airway responsiveness were taken.

The study by Moneret-Vautrin (1987) used a single blind, placebo-controlled challengeprotocol to study 30 asthmatic patients undergoing oral challenges with 2.5g MSG. The

authors did not report the MSG history of the test subjects. No specific diet control wasexercised during the course of the study. Declines in PEF were used as an indicator of a

positive response, with PEF measurements being taken hourly for 12 hours after challenge.

All treatment with corticoids was ceased 21 days prior to challenge, and treatment withtheophylline was ceased three days prior to challenge. Two out of the 30 subjects were

reported as having a positive reaction to MSG 6-10 hours after challenge.

This study has been criticised for the following reasons: the two positive reacting subjectswere not rechallenged in a double blind protocol; both subjects exhibited wandering baseline


21/36

21

PEF values during their placebo challenges, therefore differences between placebo and MSGPEF measurements would have been difficult to detect; and bronchodilator therapy was

discontinued three days before challenge, which could have led to airway instability,particularly as 7 of the 30 subjects tested were reportedly allergic to house dust.

Schwartzstein et al(1987) studied a total of 12 mildly asthmatic subjects using a doubleblind, placebo controlled protocol. The study was an outpatient study so the authors were not

able to supervise diets with respect to MSG content. Six of the subjects did not requireasthma medication and the other six were able to discontinue their medication for 12 hours

without any change in lung function measurement. One subject had a positive history of

asthmatic attacks following ingestion of a Chinese meal. Challenges were done with 1.5gMSG and used forced expiratory volume in one-second (FEV1) measurements plus the

occurrence of asthma symptoms as indicators of whether an asthma attack had occurred.FEV1 is an effort-independent measurement, which measures how much air can be blown out

in one second of a forced manoeuvre. FEV1 measurements were taken hourly for 4 hours

after challenges with placebo or MSG. No subjects in the study were reported as having an

MSG-induced asthma attack.

The criticisms of this study include: only one subject with a positive MSG history was

recruited; the total study population was considered too small; the largest challenge dose used

may have been too low (1.5g, compared to the 2.5g used in previous studies); lack of dietarysupervision; and lung function measurements were only performed for up to 4 hours after

challenge, compared to 12 hours for previous studies.

Germano et al(1991) studied 13 non-asthmatics and 30 asthmatics using a single blind oral

challenge protocol with MSG administered in capsules containing increasing doses at 30-minute intervals for a total dose of 7.6g. Two of the subjects had a positive history ofreacting to food containing MSG. Subjects were maintained on their asthma medications

throughout the study. The study was an outpatient study and it is not known if any diet

control was used. A positive reaction was defined as >20% fall in FEV1 following MSGchallenge. One of the subjects exhibited a significant drop in FEV1 following MSG

challenge. This subject was rechallenged using a double blind placebo controlled protocolwith no change in FEV1 being observed.

This study has been criticised for the following reasons: only 2 of the subjects used in thestudy had a history of bronchoconstriction after a Chinese restaurant meal; and the study was

only reported in abstract form and therefore few experimental details are available.

Altman et al(1994) recruited 47 subjects for a study using a double blind placebo controlled

protocol, although only eight of these were reported as having asthma. It is unknownwhether the subjects were subject to any diet control during the course of the study or

whether any changes were made to the asthma medications of any of the asthmatic subjects.The study was conducted in two phases. In phase I, three doses of MSG (1.5g, 3.0g, 6.0g)

and three placebo does in a liquid vehicle were administered after an overnight fast in random

order on different days. The subject recorded symptoms in a 24-hour diet/symptom diary.Phase II repeated the challenge using self-administered capsules at home. Eleven out of the

26 people who completed Phase I reported symptoms after both MSG and placebo, and twoafter placebo only. Six reported no symptoms after any dose and seven after MSG only. In

two of these cases, symptoms were reported at 3g but not at 6g. Ten out of the 16 subjects,who completed Phase II, reported no symptoms after any dose. Symptoms that were reported


22/36

22

were of short duration and did not affect daily activities. None of the subjects that hadasthma were reported as having any asthmatic symptoms following MSG challenge.

This study has been criticised for the following reasons: the study was reported in abstract

form only and therefore contains very little experimental detail; only a small number of

asthmatic subjects were used and it is not known if any of these had a history of reacting toMSG; self-reported asthma symptoms were used rather than objective measures of asthma

status; the study was funded in part by the International Glutamate Technical Committee andtherefore has been considered by some to not be independent.

The Hodge et al(1996) study was designed to compare two different methods of testing forasthma reactions, however one of the substances used was MSG. A total of 11 asthmatic

subjects were tested using a double blind placebo control challenge protocol. One of the twomethods being tested required subjects to comply with a specific diet. All subjects continued

to use their usual asthma medications. FEV1 measurements were taken for two hours

following each challenge. Graded doses from 1.2g up to 4.8g MSG were administered in

capsule form. One of the subjects was reported as having and MSG-induced asthma attack.

The main criticism of this study is that its main aim was not to explore MSG-induced asthma

therefore it is difficult to fully interpret the MSG results.

Woods et al(1998) undertook an outpatient study using 12 subjects with clinically

documented asthma and a perception of MSG-induced asthma. Usual bronchodilatormedications were continued and subjects complied with strict diet avoidance of MSG during

the study. A randomised, double blind, placebo-controlled challenge protocol was used with

subjects being administered with 1g and 5g MSG in capsule form (placebo used was 5glactose). After challenge, subjects were monitored using FEV1 measurements for 8 hours andthen sent home for self-monitoring for the next 4 hours using a PEF monitor. The study also

measured bronchial hyper responsiveness and soluble inflammatory markers. No immediate

or late asthmatic reactions were apparent in any of the subjects after oral challenge with 5gMSG.

This study has been criticised for the following reasons: as an outpatient study, the reliability

of the dietary program could not be supervised directly; during the last 4 hours of the post-

challenge observation period, patients were at home performing unsupervised PEFmeasurements; and the study only looked at a small number of subjects.

Woessneret al(1999) recruited 100 subjects, 30 of whom had a history of Chinese restaurantasthma attacks and the remaining 70 subjects had suspected aspirin-sensitive asthma and did

not have a perceived sensitivity to MSG. Subjects were admitted to an in-patient facility onthe day prior to commencement of the challenges and remained in the facility for the duration

of the study. The study used a single blind, placebo-controlled challenge protocol. Subjectsfollowed a low MSG diet throughout the study. FEV1 baseline measurements were taken

prior to commencement of the study. Placebo challenges (2.5g sucrose capsules) were given

in the morning and afternoon on the first day of the study followed by hourly FEV1measurements for a total of 12 hours. This was followed on the second day with MSG

challenges (2.5g capsules) if during the placebo challenge, FEV1 values varied by less than10% over the course of observation. Again, hourly FEV1 measurements were taken for a

total of 12 hours. The criteria used for a presumptive MSG-induced asthma attack was a 20%decline in FEV1 values from baseline with or without accompanying symptoms. If there was


23/36

23

a 20% drop in FEV1 value, serum tryptase levels were determined and the subject underwenttwo double blind placebo-controlled MSG challenges on days 3 and 4. Only 1 of the 30

subjects with a history of asthma attacks following a Chinese restaurant meal experienced a20% decline in FEV1 values during the single blind screening challenge with MSG. The

subject was without asthma symptoms throughout the MSG challenge and serum tryptase

levels were normal. Subsequent double blind placebo-controlled MSG challenges in replicatewere negative, with the post-MSG changes in FEV1 values of less than 1%. No other

subjects had a significant fall in FEV1 value or the development of asthma symptoms duringthe MSG challenge. The mean change in FEV1 with MSG challenge was no different from

that of placebo challenge. For 15 of the 30 subjects who had previously perceived

themselves to be MSG sensitive, causes other than MSG were identified as the trigger factorfor their asthma attacks following a Chinese restaurant meal.

The criticisms of this study are that it was partly funded by the International Glutamate

Technical Committee and that details of the low MSG diet were not reported.

Discussion

Virtually all of the studies reviewed contained design flaws of some description. The most

consistent problem with studies is the continuation versus discontinuation of asthma

medication. While the continuation of medication could potentially prevent the triggering ofan MSG-induced asthmatic attack, the discontinuation of the medication could result in the

occurrence of a spontaneous asthmatic attack, which could incorrectly be attributed to MSG.Notwithstanding this, the FASEB review found that the report of Allen et al(1987) was a

reasonably well-designed scientific oral challenge study in asthmatic subjects that provided

evidence to support the existence of a subgroup of asthmatic responders to MSG (Raiten etal1995). The FASEB report therefore concluded that there appears to be a small subset of

people with severe unstable asthma who respond to doses of 1.5 2.5g MSG given in capsule

form without food. Others have suggested however that the selection of subjects with

unstable asthma, combined with the discontinuation of their daily asthma medication,resulted in the subjects in both the Allen et al(1987) and Moneret-Vautrin (1987) study

developing nothing other than spontaneous asthma as would be expected in patients deprivedof their essential maintenance medications (Stevenson 2000).

It is difficult to reconcile the results of the Allen et al(1987) and Moneret-Vautrin (1987)studies with those of the Woods et al(1998) and Woessneret al(1999) studies, both of which

failed to demonstrate MSG-induced asthma attacks and which were undertaken after the

FASEB review. These two studies, particularly that of Woessneret al(1999), haveaddressed many of the design flaws of earlier studies and also clearly demonstrate the

importance of double blind challenges in verifying a positive reaction. While both theGermano et al(1991) and Woessneret al(1999) studies identified individuals exhibiting a

positive reaction to MSG on single blind challenge, subsequent double blind challengeprotocols failed to reproduce the positive reactions. This type of follow-up was not done with

the earlier studies of Allen et al(1987) and Moneret-Vautrin (1987).

Conclusion

On balance, and taking into account the design and methodological flaws evident in many of

the studies as well as the conflicting results that have been produced, the evidence for MSG-induced asthma attacks is inconclusive. More recent studies suggest MSG may not be a


24/36

24

significant trigger factor. Further challenge studies, conducted along the lines of theWoessneret al(1999) study, would be useful to help resolve the ongoing debate about

whether MSG is a trigger factor for asthmatic attacks.

7.2.2 MSG as the causative agent of CRS

A number of published case reports, seemingly prompted by the appearance of the first case

report of CRS (Kwok 1968), have suggested a causative role for MSG in CRS (Schaumburg1968, Menken 1968, Beron 1968, Migden 1968, Rath 1968, Rose 1968). Since then a large

number of clinical studies have been conducted but have produced conflicting results. Some

studies have reported significant increases in symptoms after ingestion of MSG (e.g.Schaumburg et al1969, Rosenblum et al1971, Kenney and Tidball 1972, Gore and Salmon

1980, Yang et al1997), whereas others have not or have been more equivocal (e.g. Zanda etal1973, Kenney 1986, Wilkin 1986, Tarasoff and Kelly 1993, Geha et al2000a).

The first clinical study was conducted by Schaumburg et al(1969) who administered MSG in

a variety of vehicles such as soup, water, chicken broth and intravenously. Doses rangedfrom 1 12g, and a variety of double, single and unblinded tests were conducted. The studyfound that intravenous or oral administration of MSG could cause dose-dependent symptoms

in nearly all six subjects tested.

Rosenblum et al(1971) conducted both single and double blind studies with 99 human

volunteers using doses up to 12g MSG in water. Symptoms of light-headedness and tightnessin the face appeared significantly more often in the MSG group than in the control but no

subjects reported the characteristic triad of CRS symptoms. Measurements of blood pressure,

pulse and serum chemistries were not significantly different between reactors and non-reactors.

Kenney and Tidball (1972) used an initial group of 77 subjects who they challenged with 5g

MSG in tomato juice to identify MSG-sensitive individuals. Twenty-two of the 25 whoreacted to this dose were then challenged with doses ranging from 1 4g MSG. A dose-

response relationship in the symptoms of stiffness/tightness in the face and neck wasobserved and a less clearly defined dose-response in the symptoms of tingling, pressure and

warmth was also observed. There was a threshold dose of 2 3g before any symptoms

occurred but at the 1g dose level, a greater number of subjects reported adverse reactions toplacebo than to MSG. Plasma glutamate levels were monitored in the subjects and while it

was found that the rise in plasma glutamate was significant after ingestion of MSG, there was

no significant difference in the level of plasma glutamate between reactors and non-reactors.

Zanda et al(1973) administered 3g MSG in a double blind study to 73 healthy subjects. Allsubjects were evaluated for subjective (e.g. burning sensation, nausea, headache) as well as

objective (e.g. pulse rate, arterial blood pressure) changes. No differences in symptomologywere observed between groups.

Gore and Salmon (1980) conducted a double-blind study with 55 subjects with no priorhistory of CRS. Subjects ingested three different doses of MSG (1.5, 3 and 6g) or a placebo

in 150ml cold water after an overnight fast. Nine of the subjects reacted to MSG, two reactedto placebo and three reacted to both. Reactions to MSG (abdominal cramps, headache,

nausea, and hypersalivation) were statistically more frequent but were not dose-related andwere not typical of CRS.


25/36

25

Kenney (1986) used a double blind placebo controlled protocol to challenge six subjects who

considered themselves to be MSG sensitive. The MSG was administered in a drink vehicleformulated to mask the taste of MSG. Challenges were done using 6g MSG. Four of the six

subjects did not react to either MSG or placebo, and the remaining two reacted to both MSG

and placebo. Of the subjects who reacted, one reported tingling of hands and warmth behindthe ears after both MSG and placebo and the other subject experienced tightness of the face

after ingesting either substance.

Wilkin (1986) undertook a study of flushing in 24 subjects, 18 of who had a history of

flushing symptoms after eating Chinese foods. Subjects were challenged with 3 18.5gMSG and none of the subjects reported flushing symptoms.

Tarasoff and Kelly (1993) undertook a double blind study with 71 healthy subjects using

doses of 1.5, 3.0 and 3.15g MSG. The MSG was administered in capsules as well as in

specially formulated drinks that masked the taste of MSG. Most of the subjects tested

reported no reactions to either placebo or MSG. Of the subjects that did react, the symptomsreported did not occur at a significantly higher rate than those elicited by placebo.

Yang et al(1997) conducted a double blind, placebo-controlled challenge study with 61 self-

identified MSG-sensitive subjects. Subjects were enrolled in the study on the basis that theyexperienced, within 3 hours of a meal alleged to have contained MSG, two or more of the

symptoms typically associated with CRS. Symptoms identified by subjects prior to the studywere designated as index symptoms. All non-index symptoms noted after challenge were

designated as other symptoms. All subjects underwent an initial challenge in which they

ingested on an empty stomach 5g of MSG (dissolved in 200ml of a strongly citrus tastingbeverage, containing sucrose as a sweetening agent) or placebo (same beverage withoutMSG) in random order on different days. Subjects who responded only to a single test agent

then underwent rechallenge in random sequence in a double-blind fashion with placebo and

1.25, 2.5 and 5g MSG. A positive response was defined as the reproduction of2 of thespecific symptoms in a subject, ascertained on pre-challenge interview. Of the 61 subjects

who entered the study, 18 responded to neither MSG nor placebo, 6 to both, 15 to placeboand 22 to MSG. The rates of reaction were not statistically significant with a greater than

expected rate of reactivity to placebo. More symptoms were reported after ingestion of MSG

(104 index, 105 other) than placebo (79 index, 76 other) however the differences were notstatistically significant, although a feeling of flushing occurred at a statistically increased

frequency after MSG ingestion compared with after placebo. The study demonstrated that

the sequence of administration had introduced a bias into the study, with an unbalancedresponse to placebo being recorded. Fourteen of the 31 subjects who received placebo first

responded positively compared with only 7 of 30 when placebo was administered second. Incontrast, identical numbers responded to MSG administered either first or second. The

rechallenge phase maintained the double-blind state. Of the original 37 uni-responders, onlyone declined rechallenge, which was done in random sequence with placebo and MSG at

doses of 1.25, 2.5 and 5g. Analysis of rechallenge data revealed no effect of sequence of

administration on the responses. Results showed that response to placebo was still aconfounding part of the data, however analysis of the response found that frequency and

severity of responses increased with increasing doses of MSG. Rechallenge also revealed anapparent threshold dose for reactivity of 2.5g MSG. Headache, muscle tightness, general

weakness and flushing occurred more frequently after MSG than placebo ingestion. Theauthors concluded that these results support the conclusions of the FASEB review and


26/36

26

suggest that sensitivity to MSG exists, at least in the clinical setting described and ischaracterised by unpleasant reactions such as numbness, tingling, headache, muscle tightness,

general weakness, and flushing.

Geha et al(2000a) conducted a multi-centre, double blind placebo-controlled challenge study

of 130 subjects to analyse the response of subjects who report symptoms from ingestingMSG. This study was conducted according to the criteria established by FASEB for the

confirmation of MSG symptom complex, that is, three double blind, placebo-controlledchallenges on separate occasions must reproduce symptoms with the ingestion of MSG and

produce no response with placebo (Raiten et al1995). In 3 of the 4 protocols, MSG was

administered without food in a 200ml citrus-flavoured beverage. A positive response wasscored if the subject reported 2 or more symptoms from a list of 10 symptoms (general

weakness, muscle tightness, muscle twitching, flushing, sweating, burning sensation,headache-migraine, chest pain, palpitations, numbness-tingling) reported to occur after

ingestion of MSG-containing foods within 2 hours. In protocol A, 130 self-selected

reportedly MSG-reactive volunteers were challenged with 5g of MSG and with placebo on

separate days (days 1 and 2). Of the 86 subjects who reacted to MSG, placebo, or both inprotocol A, 69 completed protocol B to determined whether the response was consistent anddose dependent. To further examine the consistency and reproducibility of reactions to MSG,

12 of the 19 subjects who responded to 5g of MSG but not to placebo in both protocols A and

B were given, in protocol C, 2 challenges, each consisting of 5g of MSG versus placebo.

Of 130 subjects in protocol A, 50 (38.5%) responded to MSG only, 17 (13.1%) responded toplacebo only, and 19 (14.6%) responded to both. Challenge with increasing doses of MSG in

protocol B was associated with increased response rates. Only half (n = 19) of 37 subjects

who reacted to 5g of MSG but not to placebo in protocol A reacted similarly in protocol B,suggesting inconsistency in the response. Two of the 19 subjects responded in bothchallenges to MSG but not placebo in protocol C; however their symptoms were not

reproducible in protocols A through C. These two subjects were challenged in protocol D 3

times with placebo and 3 times with 5g of MSG in the presence of food. Both responded toonly one of the MSG challenges in protocol D and in neither case were the symptoms the

same as those reported in the previous protocols.

The authors concluded that large doses of MSG given without food may elicit more

symptoms than a placebo in individuals who believe they react adversely to MSG. However,they noted that neither persistent nor serious effects from MSG ingestion were observed, and

frequency of responses was low. Moreover, the responses reported were inconsistent and

were not reproducible, particularly when MSG was given with food.

Discussion

One of the difficulties in studying adverse reactions to MSG is that the majority of reportedsymptoms (e.g. headache, numbness, tingling, muscle tightness) are subjective and there are

no objective clinical measures associated with the wide variety of symptoms described.

Because of this a placebo response would be expected to play a significant role in many ofthe reactions observed and this has made it hard to interpret the significance of any responses

to MSG. Furthermore, many of the studies that have attempted to establish if a link existsbetween MSG and CRS have suffered from a number of methodological flaws (Tarasoff and

Kelly 1993, Taliaferro 1995, Yang et al1997, Samuels 1999, Geha et al2000a). Many of theprevious studies were unblinded or single blinded, or if they were double blinded did not take


27/36

27

any steps to disguise the taste of MSG. Often too few subjects were used and in many studiesthe results are confounded by symptom suggestion, where subjects have been notified of

possible symptoms prior to testing. Other problems relate to the use of subjects that have noprevious history of CRS or sensitivity to MSG, and use of inappropriate placebos.

While these studies have largely failed to demonstrate a causal association between MSG andCRS, what they have demonstrated is that symptoms resembling those of CRS may be

provoked in a clinical setting in some individuals by the administration of large doses ofMSG without food.

This was largely the conclusion drawn by the FASEB Expert Panel, who although consideredthat causality had not been established, did consider there was sufficient evidence to support

the existence of a subgroup of the general population of otherwise healthy individuals whomay respond to large doses (3g) of MSG under specific conditions (i.e., an oral bolus dose

in the absence of food) (Raiten et al1995). The reactions were categorised by the Expert

Panel as acute, temporary and self-limited and the mechanism of these reactions are

unknown.

Only two further studies (Yang et al1997, Geha et al2000a) have been conducted since the

FASEB review. Both these studies have been arguably better conducted than many of the

previous studies. Both studies were double-blinded, used a liquid rather than capsule vehicleand controlled for the taste of MSG, used subjects self-identified as MSG sensitive, used an

appropriate placebo, and, in addition, the Geha et al(2000a) study used three separate doubleblind challenges as recommended by the FASEB Expert Panel. Both studies indicate that

MSG, given in relatively large doses without food, will elicit a higher frequency of symptoms

than placebo in certain individuals who consider themselves sensitive to MSG. These resultsappear to be consistent with the conclusions drawn by the FASEB review. The results of theGeha et al(2000a) study also suggest that in the presence of food the frequency of response

will be reduced, as would be expected from pharmacokinetic studies with MSG.

An interesting observation that can be made from the various studies conducted to date is that

it appears not all individuals who report as MSG-sensitive react to MSG in double blindchallenges, suggesting that they may not be sensitive to MSG at all. This highlights the

importance of having suspected sensitivities appropriately investigated as many individuals

may be unnecessarily avoiding MSG in their diets.

Further studies would be helpful, firstly to ascertain the true prevalence of reactions to MSG

in the general population, secondly to investigate how the ingestion of MSG with food islikely to affect any adverse response and thirdly to ascertain the mechanism(s) behind the

reactions observed. The elucidation of a physiological mechanism behind CRS is likely tolead to the development of more objective clinical measures for the response and thus make

challenge studies less open to residual confounding.

Conclusion

The evidence suggests that ingestion of large amounts (3g) of MSG may be responsible for

causing symptoms similar to CRS in a small subset of individuals. These symptoms,although unpleasant, are neither persistent nor serious and appear more likely to occur when

MSG is ingested in the absence of food. As MSG would always be consumed in the presenceof food, an important question that remains unanswered by the scientific literature is what


28/36

28

effect consumption with food would have on the incidence and severity of symptoms. Thepharmacokinetic evidence suggests food, particularly carbohydrate, would have an

attenuating affect.


29/36

29

REFERENCES

Abramson, M.J., Kutin, J.J. and Bowes, G. (1992). The prevalence of asthma in Victorianadults.Aust. NZ. J. Med. 22: 358 363.

Abramson, M., Kutin, J., Rosier, M. and Bowes, G. (1995). Morbidity, medication andtrigger factors in a community sample of adults with asthma.Med. J. Aust. 162: 78 81.

Airoldi, L., Bizzi, A., Salmona, M. and Garattini, S. (1979). Attempts to establish the safety

margin for neurotoxicity of monosodium glutamate. In: Glutamic Acid: Advances in

Biochemistry (Filer, L.J., Garattini, S., Kare, M.R., Reynolds, W.A. and Wurtman, R.J., eds),Raven Press, New York, NY, pp 321 331.

Allen, D.H. and Baker, G.J. (1981). Chinese restaurant asthma [Letter].New Engl. J. Med.

278: 796.

Allen, D.H., Delohery, J. and Baker, G. (1987). Monosodium L-glutamate-induced asthma. J.Allergy Clin. Immunol. 80: 530 537.

Altman, D.R., Fitzgerald, T. and Chiaramonte, L.T. (1994). Double-blind placebo-controlled

challenge (DBPCC) of persons reporting adverse reactions to monosodium glutamate (MSG).J. Allergy Clin. Immunol.93: 303.

Anderson, J.A. (1996). Allergic reactions to foods. Crit. Rev. Food Sci. Nut. 36 (S): S19

S38.

ANZFA (1997).Identification of food and food components causing frequent and severeadverse reactions. Report of the Australia New Zealand Food Authority Expert Panel on

Adverse Reactions to Food. Australia New Zealand Food Authority, Canberra.

Battaglia, F.C. (2000). Glutamine and glutamate exchange between the fetal liver and the

placenta. In:International Symposium on Glutamate, Proceedings of the symposium heldOctober, 1998 in Bergami, Italy. J. Nutr. 130 (Suppl.): 974S 977S.

Beron, E.K. (1968). Chinese-restaurant syndrome [Letter].N. Engl. J. Med.278: 1123.

Bizzi, A., Veneroni, E., Salmona, M. and Garattini, S. (1977). Kinetics of monosodium

glutamate in relation to its neurotoxicity. Toxicol. Lett. 1: 123 130.

Bock, S.A. (1987). Prospective appraisal of complaints of adverse reactions to foods inchildren during the first three years of life.Pediatrics79: 683 688.

Bock, S.A. and Aitkins, F.M. (1990). Patterns of food hypersensitivity during sixteen years of

double-blind, placebo-controlled food challenges. J. Pediatr. 117: 561 567.

Brosnan, J.T. (2000). Glutamate, at the interface between amino acid and carbohydrate

metabolism. In:International Symposium on Glutamate, Proceedings of the symposium heldOctober, 1998 in Bergami, Italy. J. Nutr. 130 (Suppl.): 988S 990S.


30/36

30

Burks, A.W. and Sampson, H. (1993). Food allergies in children. Curr. Prob. Paed. 23: 230 252.

Caccia, S., Garattini, S., Ghezzi, P. and Zanini, M.G. (1982). Plasma and brain levels of

glutamate and pyroglutamate after oral monosodium glutamate to rats. Toxicol. Lett. 10: 169

175.

Chin, K.W., Garriga, M.M. and Metcalfe, D.D. (1989). The histamine content of orientalfoods.Food Chem. Toxicol. 27: 283 287.

Dawson, K.P., Ford, R.P.K., and Mogridge, N. (1990). Childhood asthma: What do parentsadd or avoid in their childrens diet?NZ Med. J. 103: 239 240.

FASEB (1995). Analysis of Adverse Reactions to Monosodium Glutamate (MSG), Report,Life Sciences Research Office, Federation of American Societies for Experimental Biology,

Washington, DC.

Folkers, K., Shizukuishi, S., Willis, R., Scudder, S.L., Takemura, K. and Longenecker, J.B.(1984). The biochemistry of vitamin B6 is basic t

Date post:	06-Apr-2018
Category:	Documents
Upload:	riniwinnoto
View:	217 times
Download:	0 times

MSG Technical Report

Documents